Clim. Past Discuss., 8, C1334–C1338, 2012
www.clim-past-discuss.net/8/C1334/2012/
© Author(s) 2012. This work is distributed under
the Creative Commons Attribute 3.0 License.
Climate
of the Past
Discussions
CPD
8, C1334–C1338, 2012
Interactive
Comment
Interactive comment on “Palynological evidence
for gradual vegetation and climate changes during
the “African Humid Period” termination at 13◦ N
from a Mega-Lake Chad sedimentary sequence”
by P. G. C. Amaral et al.
U Salzmann (Referee)
[email protected]
Received and published: 23 August 2012
Full Screen / Esc
GENERAL COMMENTS
Printer-friendly Version
The paper presents a reconstruction of middle Holocene vegetation at Lake Chad
based on a sediment sequence covering a relatively short period of ca. 1700 yrs. The
interpretation of fossil spectra is accompanied by a thorough study of modern pollen
rain for calibration as well as biomization and quantitative estimate of Holocene mean
annual precipitation.
C1334
Interactive Discussion
Discussion Paper
Although presenting only a short pollen sequence, Amaral et al.’s study is an important
contribution to a better understanding of the termination of the “African Humid Period”.
The interpretation of pollen spectra is detailed and based on a profound knowledge of
African plant ecology. The only concern I have is related to the quantitative estimates
of mean annual precipitation. The estimates are presented with very large error bars
resulting in mid Holocene precipitation ranges of 400-1500 mm/a. This precipitation
range comprises the present rainfall of multiple West African vegetation zones, including rain forest, savanna and semi-deserts. Given the large uncertainty, does it make
sense at all to use absolute rainfall estimates, or should this approach better used for
identifying trends only (wetter to drier)?
CPD
8, C1334–C1338, 2012
Interactive
Comment
Estimating rainfall from fossil pollen data in semi-arid and arid regions is strongly hampered by the occurrence of extrazonal vegetation (i.e. vegetation that occurs outside
of its main distribution area by occupying habitats with favourable environmental (e.g.
edaphic) conditions, such as swamps and lake shores). I do not fully understand from
the paper how the authors solved this issue. They mention that swamp taxa have been
excluded. Does this include Uapaca, Alchornea and Szygium?
Amaral et al. compare their record with published pollen spectra from the neighbouring Manga Grasslands(e.g. Salzmann & Waller 1998) which also show high Uapaca
percentages. Amaral et al. suggest that the Uapaca-rich savannas of Togo might be
a potential modern analogue vegetation. In the Manga Grasslands we used multiple
pollen records to distinguish between local (lake) and regional (surrounding savannah)
vegetation. We could unambiguously identify Uapaca as a swamp forest element and
I therefore disagree with this reconstruction of a zonal Uapaca dominated Guinean
savannah. At least for the Manga Grasslands, such swamp taxa should not be used
for estimating regional rainfall. Please explain why it is assumed that at Lake Chad
potential swamp taxa such as Uapaca,Szygium and Alchornea are part of a savanna.
Which elements in the Lake Chad pollen profile are representative of a vegetation fringing the large shores and swampy areas of the mega lake? How has the habitat been
C1335
Full Screen / Esc
Printer-friendly Version
Interactive Discussion
Discussion Paper
identified?
My concern regarding rainfall estimates from fossil data is an additional comment rather
than a major criticism of the general interpretation of pollen data, which I fully support.
However I would appreciate if extrazonal vegetation and the inherent problem of estimating rainfall in semi-arid regions could be discussed in more detail.
CPD
8, C1334–C1338, 2012
Interactive
Comment
SPECIFIC COMMENTS
Abstract
Abstract is a bit long and could be shortened. 2322/15: Not sure what “extra-local environment” means. Can it be replaced by “zonal” or “regional”? 2322/28-20: A circular
argument: Palynology and reconstructed palaeovegetation has been used to estimate
biomes and annual rainfall. Therefore the reconstructed “vegetation distribution” cannot be “supported..by biome reconstructions and mean annual precipitation estimates”,
as indicated in the abstract.
Introduction
2323/13: Please delete “ca. 6000-5500 yrs ago” or provide reference, as the termination of the AHP and establishment of the modern hyper-arid Sahara is still under
discussion.
2323/22: For African lake data sets please cite:
Full Screen / Esc
Hoelzmann, P., D. Jolly, S. P. Harrison, F. Laarif, R. Bonnefille, H.-J. Pachur, and
TEMPO (1998), Mid-Holocene land-surface conditions in northern Africa and the Arabian Peninsula: A data set for the analysis of biogeophysical feedbacks in the climate
system, Global Biogeochem. Cycles, 12(1), 35–51,
Printer-friendly Version
Lézine, A.-M., C. Hély, C. Grenier, P. Braconnot, and G. Krinner. 2011. Sahara and
Sahel vulnerability to climate changes, lessons from Holocene hydrological data. Quaternary Science Reviews.
Discussion Paper
C1336
Interactive Discussion
3224/8: Please reword “local recycling of precipitation”
CPD
2326/2: There are more than four records for the Manga Grasslands, which might be
of interest for this paper. See also:
8, C1334–C1338, 2012
Waller, Martyn P., Street-Perrott, F. Alayne and Wang, Hongya (2007) Holocene vegetation history of the Sahel: pollen, sedimentological and geochemical data from Jikariya
Lake, north-eastern Nigeria. Journal Of Biogeography, 34(9), pp. 1575-1590
Interactive
Comment
Street-Perrott, FA, Holmes, JA, Waller, MP, Allen, MJ, Barber, NGH, Fothergill, PA,
Harkness, DD, Ivanovich, M, Kroon, D and Perrott, RA (2000) Drought and dust deposition in the West African Sahel: A 5500-year record from Kajemarum Oasis, northeastern Nigeria.The Holocene, 10(3), pp. 293-302.
2326/10: The core is not continuous if the upper part is truncated (2329/3).
Material and Methods
2328/19 Please show sedimentology (legend is missing) in Fig.3
2330/17 and 2331/8 There are many trees and shrubs which grow in both wet savannah zones/forests and semi-arid and arid areas with favourable edaphic condition,
such as swamps (see “General Comments” on extrazonal vegetation). Please clarify
how “non-edaphic pollen taxa” and “swamp trees” have been identified and removed?
Did this include Uapaca and Syzygium?
Full Screen / Esc
Results
2332/27 Please provide in addition to the summary pollen diagram the full pollen record
with all taxa; either in the main text or supplement.
2333/4 Fig 5 would be easier to read if latitudes were indicated for each site.
Interpretation and Discussion
Printer-friendly Version
Interactive Discussion
Discussion Paper
2340 4-8 Please reword, as the sentence is not in line with the authors interpretation
C1337
(see Salzmann and Waller, 1998). Instead we showed with our multiple pollen records
that Uapaca migrated into the Sahel and was present in the Manga Grassland AND
colonised the dune depression. The occurrence of Uapaca is not an either/or question
as suggested in this manuscript.
2341/19 A Holocene “Guineo-Congolian lowland rain forest” at Lake Tilla is a (re)interpretation which is not supported by Salzmann et al. 2002. Please delete.
2343/16 long distance transport instead of “extra-regional” ?
CPD
8, C1334–C1338, 2012
Interactive
Comment
2343/16 see also Waller et al. 2007
Conclusions
2346/17-19 Circular argument (see abstract)
TECHNICAL CORRECTIONS
Check abbreviations for yr and yrs
2323/23 Kuper et Kropelin
3224/20 Regardless. . .
2328/5-6 Too many full stops
2334/3 “older/younger than” instead of “after”
Full Screen / Esc
2337/9 Grammar error
2341/12 Schulz instead of Shlulz (also check typo in respective reference “Hydrobiologia”)
Printer-friendly Version
Interactive Discussion
Interactive comment on Clim. Past Discuss., 8, 2321, 2012.
Discussion Paper
C1338